Skip to main content
. 2021 Feb 24;12:1264. doi: 10.1038/s41467-021-21541-5

Fig. 3. Pairwise comparison across ten different reference panels.

Fig. 3

We collected ten reference panels of differing ethnicities and sizes. We then tested every possible pair by assigning one panel as a reference and another as a target, which comprised 88 tests after excluding overlapping sample pairs. We compared the full version of CookHLA to CookHLA-HapMap (upgraded engine, with HapMap genetic map). The dotted line indicates where the two methods’ imputation accuracies are equal.